Radeon RX 9070 vs 625
Aggregate performance score
We've compared Radeon 625 with Radeon RX 9070, including specs and performance data.
9070 outperforms 625 by a whopping 2256% based on our aggregate benchmark results.
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 850 | 41 |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | no data | 65.29 |
| Power efficiency | 3.91 | 20.99 |
| Architecture | GCN 3.0 (2014−2019) | RDNA 4.0 (2025) |
| GPU code name | Polaris 24 | Navi 48 |
| Market segment | Laptop | Desktop |
| Release date | 13 May 2019 (6 years ago) | 6 March 2025 (less than a year ago) |
| Launch price (MSRP) | no data | $549 |
Cost-effectiveness evaluation
The higher the ratio, the better. We use the manufacturer's recommended prices.
Performance to price scatter graph
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.
| Pipelines / CUDA cores | 384 | 3584 |
| Core clock speed | 730 MHz | 1330 MHz |
| Boost clock speed | 1024 MHz | 2520 MHz |
| Number of transistors | 1,550 million | 53,900 million |
| Manufacturing process technology | 28 nm | 5 nm |
| Power consumption (TDP) | 50 Watt | 220 Watt |
| Texture fill rate | 24.58 | 564.5 |
| Floating-point processing power | 0.7864 TFLOPS | 36.13 TFLOPS |
| ROPs | 8 | 128 |
| TMUs | 24 | 224 |
| Tensor Cores | no data | 112 |
| Ray Tracing Cores | no data | 56 |
| L0 Cache | no data | 896 KB |
| L1 Cache | 96 KB | no data |
| L2 Cache | 128 KB | 8 MB |
| L3 Cache | no data | 64 MB |
Form factor & compatibility
Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).
| Interface | PCIe 3.0 x8 | PCIe 5.0 x16 |
| Length | no data | 267 mm |
| Width | no data | 2-slot |
| Supplementary power connectors | None | 2x 8-pin |
VRAM capacity and type
Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.
| Memory type | DDR3 | GDDR6 |
| Maximum RAM amount | 2 GB | 16 GB |
| Memory bus width | 64 Bit | 256 Bit |
| Memory clock speed | 900 MHz | 2518 MHz |
| Memory bandwidth | 14.4 GB/s | 644.6 GB/s |
| Shared memory | - | - |
| Resizable BAR | - | + |
Connectivity and outputs
This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.
| Display Connectors | No outputs | 1x HDMI 2.1b, 3x DisplayPort 2.1a |
| HDMI | - | + |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 (12_0) | 12 Ultimate (12_2) |
| Shader Model | 6.3 | 6.8 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.0 | 2.2 |
| Vulkan | 1.2.131 | 1.3 |
| DLSS | - | + |
Synthetic benchmarks
Non-gaming benchmark results comparison. The combined score is measured on a 0-100 point scale.
Combined synthetic benchmark score
This is our combined benchmark score.
Passmark
This is the most ubiquitous GPU benchmark. It gives the graphics card a thorough evaluation under various types of load, providing four separate benchmarks for Direct3D versions 9, 10, 11 and 12 (the last being done in 4K resolution if possible), and few more tests engaging DirectCompute capabilities.
Gaming performance
Let's see how good the compared graphics cards are for gaming. Particular gaming benchmark results are measured in FPS.
Average FPS across all PC games
Here are the average frames per second in a large set of popular games across different resolutions:
| Full HD | 9−10
−2267%
| 213
+2267%
|
| 1440p | 5−6
−2280%
| 119
+2280%
|
| 4K | 3−4
−2367%
| 74
+2367%
|
Cost per frame, $
| 1080p | no data | 2.58 |
| 1440p | no data | 4.61 |
| 4K | no data | 7.42 |
FPS performance in popular games
Full HD
Low
| Counter-Strike 2 | 7−8
−4157%
|
290−300
+4157%
|
| Cyberpunk 2077 | 5−6
−2940%
|
150−160
+2940%
|
| Hogwarts Legacy | 8−9
−3200%
|
264
+3200%
|
Full HD
Medium
| Battlefield 5 | 8−9
−2013%
|
160−170
+2013%
|
| Counter-Strike 2 | 7−8
−4157%
|
290−300
+4157%
|
| Cyberpunk 2077 | 5−6
−2940%
|
150−160
+2940%
|
| Far Cry 5 | 7−8
−4171%
|
299
+4171%
|
| Fortnite | 12−14
−2162%
|
290−300
+2162%
|
| Forza Horizon 4 | 12−14
−1738%
|
230−240
+1738%
|
| Forza Horizon 5 | 6−7
−2983%
|
180−190
+2983%
|
| Hogwarts Legacy | 8−9
−2825%
|
234
+2825%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 12−14
−1246%
|
170−180
+1246%
|
| Valorant | 40−45
−670%
|
300−350
+670%
|
Full HD
High
| Battlefield 5 | 8−9
−2013%
|
160−170
+2013%
|
| Counter-Strike 2 | 7−8
−4157%
|
290−300
+4157%
|
| Counter-Strike: Global Offensive | 50−55
−456%
|
270−280
+456%
|
| Cyberpunk 2077 | 5−6
−2940%
|
150−160
+2940%
|
| Dota 2 | 24−27
−2208%
|
600−650
+2208%
|
| Far Cry 5 | 7−8
−4029%
|
289
+4029%
|
| Fortnite | 12−14
−2162%
|
290−300
+2162%
|
| Forza Horizon 4 | 12−14
−1738%
|
230−240
+1738%
|
| Forza Horizon 5 | 6−7
−2983%
|
180−190
+2983%
|
| Grand Theft Auto V | 6−7
−2650%
|
160−170
+2650%
|
| Hogwarts Legacy | 8−9
−2100%
|
176
+2100%
|
| Metro Exodus | 5−6
−3000%
|
150−160
+3000%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 12−14
−1246%
|
170−180
+1246%
|
| The Witcher 3: Wild Hunt | 9−10
−4767%
|
438
+4767%
|
| Valorant | 40−45
−670%
|
300−350
+670%
|
Full HD
Ultra
| Battlefield 5 | 8−9
−2013%
|
160−170
+2013%
|
| Cyberpunk 2077 | 5−6
−2940%
|
150−160
+2940%
|
| Dota 2 | 24−27
−2208%
|
600−650
+2208%
|
| Far Cry 5 | 7−8
−3814%
|
274
+3814%
|
| Forza Horizon 4 | 12−14
−1738%
|
230−240
+1738%
|
| Hogwarts Legacy | 8−9
−1563%
|
133
+1563%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 12−14
−1246%
|
170−180
+1246%
|
| The Witcher 3: Wild Hunt | 9−10
−2622%
|
245
+2622%
|
| Valorant | 40−45
−670%
|
300−350
+670%
|
Full HD
Epic
| Fortnite | 12−14
−2162%
|
290−300
+2162%
|
1440p
High
| Counter-Strike 2 | 6−7
−2933%
|
180−190
+2933%
|
| Counter-Strike: Global Offensive | 18−20
−2363%
|
450−500
+2363%
|
| Grand Theft Auto V | 1−2
−12900%
|
130−140
+12900%
|
| Metro Exodus | 0−1 | 100−110 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 21−24
−661%
|
170−180
+661%
|
| Valorant | 21−24
−1674%
|
400−450
+1674%
|
1440p
Ultra
| Cyberpunk 2077 | 2−3
−4150%
|
85−90
+4150%
|
| Far Cry 5 | 4−5
−6000%
|
244
+6000%
|
| Forza Horizon 4 | 6−7
−3283%
|
200−210
+3283%
|
| Hogwarts Legacy | 3−4
−3367%
|
104
+3367%
|
| The Witcher 3: Wild Hunt | 3−4
−6133%
|
187
+6133%
|
1440p
Epic
| Fortnite | 5−6
−2920%
|
150−160
+2920%
|
4K
High
| Grand Theft Auto V | 16−18
−819%
|
140−150
+819%
|
| Valorant | 12−14
−2408%
|
300−350
+2408%
|
4K
Ultra
| Cyberpunk 2077 | 0−1 | 40−45 |
| Dota 2 | 7−8
−2186%
|
160−170
+2186%
|
| Far Cry 5 | 1−2
−13200%
|
133
+13200%
|
| Forza Horizon 4 | 2−3
−7550%
|
150−160
+7550%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 3−4
−3100%
|
95−100
+3100%
|
4K
Epic
| Fortnite | 3−4
−2533%
|
75−80
+2533%
|
1440p
Ultra
| Battlefield 5 | 150−160
+0%
|
150−160
+0%
|
4K
High
| Counter-Strike 2 | 80−85
+0%
|
80−85
+0%
|
| Hogwarts Legacy | 40−45
+0%
|
40−45
+0%
|
| Metro Exodus | 65−70
+0%
|
65−70
+0%
|
| The Witcher 3: Wild Hunt | 160
+0%
|
160
+0%
|
4K
Ultra
| Battlefield 5 | 100−110
+0%
|
100−110
+0%
|
| Counter-Strike 2 | 80−85
+0%
|
80−85
+0%
|
| Hogwarts Legacy | 60
+0%
|
60
+0%
|
This is how Radeon 625 and RX 9070 compete in popular games:
- RX 9070 is 2267% faster in 1080p
- RX 9070 is 2280% faster in 1440p
- RX 9070 is 2367% faster in 4K
Here's the range of performance differences observed across popular games:
- in Far Cry 5, with 4K resolution and the Ultra Preset, the RX 9070 is 13200% faster.
All in all, in popular games:
- RX 9070 performs better in 53 tests (87%)
- there's a draw in 8 tests (13%)
Pros & cons summary
| Performance score | 2.43 | 57.26 |
| Recency | 13 May 2019 | 6 March 2025 |
| Maximum RAM amount | 2 GB | 16 GB |
| Chip lithography | 28 nm | 5 nm |
| Power consumption (TDP) | 50 Watt | 220 Watt |
Radeon 625 has 340% lower power consumption.
RX 9070, on the other hand, has a 2256.4% higher aggregate performance score, an age advantage of 5 years, a 700% higher maximum VRAM amount, and a 460% more advanced lithography process.
The Radeon RX 9070 is our recommended choice as it beats the Radeon 625 in performance tests.
Be aware that Radeon 625 is a notebook graphics card while Radeon RX 9070 is a desktop one.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
